The increasing complexity of sustainability challenges and the growing adoption of circular economy models have reinforced the pivotal role of Life Cycle Assessment (LCA) as a strategic decision-support tool. However, the technical demands and resource intensity of LCA often hinder its widespread implementation across sectors. Recent advancements in generative artificial intelligence, particularly through large language models such as GPTs, offer promising avenues to enhance LCA practices by facilitating knowledge retrieval, standardizing processes, and supporting analytical reasoning. While early reflections have acknowledged this potential, existing research remains largely conceptual, with few empirically grounded frameworks assessing the operational viability of GPTs in LCA contexts. This study addresses this gap by developing a structured methodological framework aimed at identifying, selecting, and preparing GPT-based tools for their future application in LCA consultancy. A sample of GPTs relevant to sustainability domains was systematically mapped, and a standardized set of prompts was designed to simulate critical decision points throughout the LCA process. These prompts were refined through expert consultation to ensure methodological robustness and alignment with ISO standards. Preliminary findings suggest that, although GPTs vary widely in scope and depth, a subset demonstrates notable potential in supporting methodological structuring and inventory development tasks. By laying the groundwork for a rigorous and replicable evaluation protocol, this study advances the empirical integration of artificial intelligence in environmental modelling. It also offers practical insights for researchers and consultants seeking to responsibly harness generative AI to scale and strengthen life cycle thinking in the circular and digital transitions.